Deep Gradual Multi-Exposure Fusion via Recurrent Convolutional Network

Je Ho Ryu, Jong Han Kim, Jong Ok Kim

Research output: Contribution to journalArticlepeer-review

Abstract

The performance of multi-exposure image fusion (MEF) has been recently improved with deep learning techniques but there are still a couple of problems to be overcome. In this paper, we propose a novel MEF network based on recurrent neural network (RNN). Multi-exposure images have different useful information depending on their exposure levels, and in order to fuse them complementarily, we first extract the local detail and global context features of input source images, and both features are separately combined. A weight map is learned from the local features for effectively fusing according to the importance of each source image. Adopting RNN as a backbone network enables gradual fusion, where more inputs result in further improvement of the fusion gradually. Also, information can be transferred to the deeper level of the network. Experimental results show that the proposed method achieves the reduction of fusion artifacts and improves detail restoration performance, compared to conventional methods.

Original languageEnglish
JournalIEEE Access
DOIs
Publication statusAccepted/In press - 2021

Keywords

  • Brightness
  • Deep learning
  • Dilated convolution filter
  • Feature extraction
  • Fuses
  • Gradual fusion
  • Image fusion
  • Image reconstruction
  • Image restoration
  • Multi-exposure image fusion
  • Recurrent convolutional network

ASJC Scopus subject areas

  • Computer Science(all)
  • Materials Science(all)
  • Engineering(all)

Fingerprint

Dive into the research topics of 'Deep Gradual Multi-Exposure Fusion via Recurrent Convolutional Network'. Together they form a unique fingerprint.

Cite this